R. White
Tufts University,
United States
Keywords: MEMS, shear sensor, microphone, array, turbulent boundary layer, wind tunnel
Summary:
In aerodynamic flow testing, pressure and skin friction (surface shear stress) are the forcing functions that create lift, drag, and structural vibration. Knowledge of the steady and unsteady pressures and shear forces at a surface are needed for predicting and measuring total drag, flow separation, external acoustic generation, and internal acoustics due to structural vibrations. Work in our group at Tufts between 2006 and 2017, in collaboration with Spirit Aerosystems, NASA (Ames and Langley), Bombardier, and Draper, resulted in two micromachined (MEMS) array technologies targeted at subsonic wind tunnels. The first system is a 64 element microphone array on a chip, micromachined on a 1 cm^2 die in the PolyMUMPS process. The array was designed to measure the fluctuating pressures present under a turbulent boundary layer (TBL). The 0.6 mm diameter and 1.25 mm pitch of the elements gives high spatial resolution information that is difficult to obtain by other means. A unique switched architecture system is employed to reduce data acquisition channel count. The array has been applied to the measurement of turbulence spectra under flat plate TBLs at Mach numbers up to 0.6 and Reynolds numbers up to 10^7. Measurements have been in three flow facilities: Spirit Aerosystems 6” boundary layer flow duct, NASA Ames Fluid Mechanics Lab 14” Indraft tunnel, and the University of Toronto 120 cm x 80 cm low turbulence intensity tunnel. The second system is a MEMS floating element shear stress sensor array on a chip. In laminar flow testing, shear and pressure gradient sensitivity of the sensor were independently determined. It is very important to distinguish the two sensitivities if operating or calibrating in high pressure gradient flows. In wind tunnel testing, real time shear was measured under a TBL for a range of conditions. Shear stresses up to 6.5 Pa were measured, consistent with expectations based on the measured boundary layer profiles. Sensor resolution was 1 Pa/Hz^0.5 at a data rate of 3 samples/second. Orientation dependence of the sensor output was verified, demonstrating the ability to measure direction as well as magnitude of the shear stress. Direct comparison to oil film shear stress measurements were also made for a number of wall roughness scenarios. Packaging for these in-the-wall surface flow testing scenarios can be challenging, as the surface topology for the entire package should be kept within the viscous sublayer, a thickness of approximately 25 microns in these cases. We have attempted a few packaging approaches. The first approach uses CPGA packages with CNC milled epoxy fill, wirebonding, and vapor-phase Parylene coating. A second approach is a chip-in-board method using a milled PCB, laser cut stencil, semiconductor processing tape and conductive ink interconnect by syringe printing. Both of these methods achieve approximately 100 microns of total surface topology. We have made some limited attempts to apply aerosol jet printing for dielectric and interconnect as well, which will be described.